An effective connectivity model based on excitation-inhibition imbalance to classify states of the epileptogenic network

COLLAVINI S.; MARIANO FERNÁNDEZ CORAZZA; GRANADO M.; SILVIA KOCHEN; MURAVCHIK, CARLOS H

San Juan

Congreso; XXIII Congreso Argentino de Bioingeniería y XII Jornadas de Ingeniería Clínica; 2022

Sociedad Argentina de Bioingenieria

In this paper, we present an effective connectivity modelbased on the imbalance between neuronal excitation and inhibition toclassify epileptogenic network states. We used electroencephalographicintracranial signals from refractory epilepsy patients. Autoregressive-moving-average models were adjusted to these signals to estimate ef-fective connectivity. We estimated ‘in-degree’ and ‘out-degree’ patternsassociated with inhibitory and excitatory processes. These patterns wereused as variables or features in supervised classifiers to classify ictal, pre-ictal and basal states of the epileptogenic network. We obtained valuesof Area Under the Curve (AUC) larger than 0.99 in the distinction ofthese pathological brain states.